Food Chemistry 329 (2020) 127129 Contents lists available at ScienceDirect Food Chemistry journal homepage: www elsevier com/locate/foodchem Limited genotypic and geographic variability of 16-O-methylated diterpene content in Coffea arabica green beans T ⁎ Vincent Portaluria, Freddy Thomasa, , Sophie Guyadera, Eric Jamina, Benoit Bertrandb,c, Gérald S. Remaudd, Elisabetta Schievanoe, Stefano Mammie, Elena Guerciaf, Luciano Navarinif a Eurofins Analytics France, 9 rue Pierre Adolphe Bobierre, B.P. 42301, F-44323 NANTES Cedex 3, France b CIRAD, UMR IPME, F-34398 Montpellier, France c UMR IPME, Univ Montpellier, CIRAD, IRD, F-34398 Montpellier, France d EBSI Team, Interdisciplinary Chemistry: Synthesis, Analysis, Modelling (CEISAM), University of Nantes-CNRS UMR 6230, 2 rue de la Houssinière, BP 92208, F-44322 Nantes cedex 3, France e Department of Chemical Sciences, University of Padova, Via Marzolo 1, 35131 Padova, Italy f illycaffè S.p.A, Via Flavia 110, 34147 Trieste, Italy ARTICLE INFO ABSTRACT Keywords: The acknowledged marker of Robusta coffee, 16-O-methylcafestol (16-OMC), can be quantified by NMR as a 16-O-methylcafestol mixture with 16-O-methylkahweol (16-OMK), which accounts for approximately 10% of the mixture. In the 16-O-methylkahweol present study, we detected and quantified 16-O-methylated diterpenes (16-OMD) in 248 samples of green Coffea ff Co ea arabica arabica beans by NMR. We did not observe any differences between genotypes introgressed by chromosomal Green coffee fragments of Robusta and non-introgressed genotypes. Environmental effects suggesting a possible protective NMR role of 16-OMD for adaptation, as well as genotypic effects that support a high heritability of this trait were observed. Altogether, our data confirmed the presence of 16-OMD in green Arabica at a level approximately 1.5% that of a typical Robusta, endorsing the validity of 16-OMD as a marker for the presence of Robusta. 1. Introduction constituents of the unsaponifiable coffee oil fraction. They are mainly esterified with various fatty acids, and the free form is present only in Coffee is the most widely consumed beverage in the world and one small amounts (De Angelis et al., 2014). The three most important di- of the most commercialized food products. In spite of more than 120 terpenes are cafestol, kahweol and 16-O-methylcafestol (16-OMC). known botanical species (Davis, Govaerts, Bridson, & Stoffelen, 2006; They are produced only by plants of the Coffea genus. Cafestol is found Davis, Tosh, Ruch, & Fay, 2011), only two are commercially important: in both Arabica and Robusta coffee while kahweol is present in larger Coffea arabica L. (Arabica coffee) and Coffea canephora Pierre ex A. amounts in Arabica (0.1–1.0%) than in Robusta (up to 200 mg/kg) Froehner (Robusta coffee). Arabica is the most valuable coffee in the (Finotello et al., 2017). For years, 16-OMC was considered to be present trade because it produces a very fragrant, sweet, smooth, and slightly exclusively in Robusta (Bonnlander, Wünnecke, & Winterhalter, 2007; acidic beverage with a very rich and complex aroma. Robusta coffee has Speer & Kolling-Speer, 2001; Kamm et al., 2002; Kurzrock & Speer, rougher aromatic notes, is more bitter, and more astringent. The lower 2001; Oellig & Radovanovic, 2017; Pacetti, Boselli, Balzano, & Frega, price of Robusta relative to Arabica prompted the search for new rapid 2012; Speer & Kölling-Speer, 2006). This fact made 16-OMC an ex- methodologies to distinguish them to prevent economically motivated cellent authenticity marker for the presence of Robusta in coffee pro- adulterations (Everstine, Spink, & Kennedy, 2013). ducts, considering also its thermal stability. To the best of our knowl- Several approaches have been suggested to distinguish these coffee edge, only two early studies reported the presence of 16-OMC in roasted species, using many different compounds as reliable discriminants Arabica. The oldest one detected trace amounts (10 mg/kg) of 16-OMC (Finotello et al., 2017). In recent years, the coffee lipid fraction has in one Arabica sample from Guatemala (Speer & Montag, 1989), attracted interest because its components can be successfully used to whereas the other one reported a higher content but remarkably lower discern the two coffee species. than that found in Robusta (Lercker, Frega, Bocci, & Rodriguez-Estrada, Coffee diterpenes (Speer & Kolling-Speer, 2001) are the main 1995). ⁎ Corresponding author. E-mail address: FreddyThomas@eurofins.com (F. Thomas). hiips://doi.org/10.1016/j.foodchem.2020.127129 Received 12 December 2019; Received in revised form 20 May 2020; Accepted 22 May 2020 Available online 27 May 2020 0308-8146/ © 2020 Elsevier Ltd. All rights reserved. V. Portaluri, et al. Food Chemistry 329 (2020) 127129 Table 1 Genetic and geographical origin of the C. arabica samples used in the present investigation. Country Number of Introgressed (samples with Non introgressed (samples with Introgression not determined (samples samples 16-OMD > 50 ppm) 16-OMD > 50 ppm) with 16-OMD > 50 ppm) CIRAD collection Nicaragua 96 47 (2) 49 (2) Commercial South & Central 12 12 (0) America Africa 12 12 (0) Asia 5 5 (1) Oceania 1 1 (0) unknown 6 6 (0) Breedcafs project Nicaragua 80 60 (13) 20 (0) Central America 24 13 (2) 11 (0) Additional Central America 7 4 (2) 3 (0) South America 1 1 (0) Africa 3 3 (0) India 1 1 (0) Total 248 In addition to these three major diterpenes, 16-O-methylkahweol investigation on a larger number of coffee samples in the attempt to (16-OMK) deserves mention. First identified and quantified in Robusta verify the presence and discern the origin of 16-OMC and 16-OMK in coffee by Kolling-Speer, Kurzrock, and Speer (2001) and Kolling-Speer Arabica coffee. To avoid any possible influence of the roasting process, and Speer (2001), its amounts are about 10 times lower than those of screening was carried out on ground green coffee. Authentic Arabica kahweol and even less in roasted Arabica (Pacetti et al., 2012). The green coffee samples were characterized by 1D 1H NMR spectroscopy interest in 16-OMK stems from the use of the diagnostic NMR peak at according to the method proposed by Schievano et al. (2014) and 3.16–3.18 ppm to quantify coffee diterpenes (Finotello et al., 2017; Monakhova et al. (2015) to quantify 16-OMD. The diagnostic NMR Schievano, Finotello, De Angelis, Mammi, & Navarini, 2014). This peak peak was detected and quantified in many of the investigated green derives from the H21 methyl groups of both esterified 16-OMC and 16- Arabica beans. Screening was extended to a wide range of samples, OMK (cumulatively named hereafter 16-O-methylated diterpenes, 16- including several clones of F1 hybrids obtained by crossing wild OMD) (D’Amelio, De Angelis, Navarini, Schievano, & Mammi, 2013; Ethiopians accessions × introgressed cultivars of different geographical Gunning et al., 2018; Scharnhop & Winterhalter, 2009). Gunning et al. origin or different trade quality, in the case of commercial lots. NMR (2018) modified the sample preparation procedure used up to that experiments were also carried out on green coffee oil extracted from point to quantify 16-OMC by NMR (Defernez et al., 2017; Monakhova selected samples, and a preliminary UPLC-MS/MS method was devel- et al., 2015; Schievano et al., 2014) and improved the limit of detection. oped to validate the quantitative NMR data and to determine the In contrast to previous NMR studies, they observed 16-OMC and 16- amounts of 16-OMC and 16-OMK separately. OMK in Arabica coffees, although at very low levels. The content of 16- OMD in roasted Arabica is approximately 1.5% that of a typical Robusta (Finotello et al., 2017; Kurzrock & Speer, 2001). This amount may re- 2. Materials and methods present only a problem in revealing unintentional Robusta con- tamination in a 100% Arabica roasted coffee blend because undeclared 2.1. Coffee samples intentional additions of Robusta lower than 2% would hardly be eco- nomically advantageous. All 248 green C. arabica samples used in the present investigation The presence of trace amounts of 16-OMD in Arabica roasted beans are reported in Table 1 and described in detail in Table 1S. has more relevant biochemical and genetic implications. 16-OMC has To study the possible effect of introgression on the content of 16- been found in other tissues of the Arabica coffee plant (Speer & Kölling- OMD, 96 samples were analyzed from the CIRAD collection (Nicaragua) Speer, 2006), and 16-OMK has been found in leaves (Speer & Kolling- (see Table 1S, Source: CIRAD collection). We analyzed 36 additional Speer, 2001), suggesting that the biochemical machinery to synthesize samples from 15 different countries of Latin America (12), Africa (12), these diterpenes is present in the Arabica genome. A legitimate doubt Asia (5) and Oceania (1) and 6 of unknown origin (see Table 1S, Source: remains as to whether this capability has been transferred to Arabica commercial) for a wide geographical representation. Harvest periods of plants through the breeding programs implemented to confer resistance the samples were spread over several years between 2007 and 2015. to main diseases by introgression of appropriate C. canephora chromo- The samples were stored in tubes (10–15 g) in hermetically sealed somal fragments. Today, Arabica cultivars derived from C. canephora boxes with silica gel, kept in the dark. All of the Ethiopian coffee via the interspecific Timor Hybrid (a spontaneous cross between C. samples were from wild accessions. The samples from the CIRAD col- canephora × C. arabica) represent more than 40% of the Arabica trees lection represent 47 introgressed accessions and 49 non-introgressed cultivated around the world. Unfortunately, introgression via the Timor accessions. The introgression was verified with simple sequence repeat Hybrid is sometimes accompanied by a substantial drop in cup quality (SSR) markers at the CIRAD institute.